More Scheme

43
More Scheme CS 331

description

More Scheme. CS 331. Quiz. What is (car ‘((2) 3 4))? (2) What is (cdr ‘((2) (3) (4)))? ((3)(4)) What is (cons ‘2 ‘(2 3 4))? (2 2 3 4) What is the length of the list ‘(()()()())? 4 Which element does (car (cdr ‘(x y z))) extract from the list? y - PowerPoint PPT Presentation

Transcript of More Scheme

Page 1: More Scheme

More Scheme

CS 331

Page 2: More Scheme

Quiz 1. What is (car ‘((2) 3 4))?

(2)

2. What is (cdr ‘((2) (3) (4)))?((3)(4))

3. What is (cons ‘2 ‘(2 3 4))?(2 2 3 4)

4. What is the length of the list ‘(()()()())?4

5. Which element does (car (cdr ‘(x y z))) extract from the list?y

6. What do some people thing LISP stands for?Losta Insane Stupid Parenthesis

Page 3: More Scheme

Reading Input

• (read) returns whatever is input at the keyboard

• (define x (read))

55

x

55

Page 4: More Scheme

Display Output

• Use display:

• (display (+ 3 4))• (display x)• (newline) ; output newline

• Can use to add to functions for debugging or use Trace functions

Page 5: More Scheme

Trace – Shows function calls

• Choose “Pretty Big or R5RS)” as your language• Add (require (lib "trace.ss")) to the top of your

program• Add (trace functionname) or (untrace

functionname) to turn off tracing

Page 6: More Scheme

Structure of Lists

• List : a sequence of zero of more elements• May be heterogeneous

– (a 20 (20 20) (lambda (x) (+ x x)))

• List of Lists• (+ 3 4)

– A list

– An expression

• Allows expression to be built and then evaluated

Page 7: More Scheme

Tree structure of lists(a b c)

a

b

c ( )

(cons a

(cons b

(cons c ‘())

)

)

Page 8: More Scheme

List structure

(a (b) (c (d)))

a

( )b ( )

c

d ( )

Page 9: More Scheme

List storage

• Independent of allocation schemes

• Familiarity is helpful for assessment

• Cons is the constructor– Allocates a single cell

head

tail

Page 10: More Scheme

List structure

(a (b) (c (d)))

( )

a

b ( ) c d ( )

Page 11: More Scheme

Notion of Equality

• Eq? – Checks if the two pointers are the same

• Equal? – Checks if the two arguments are lists with

“equal” elements.– Recursive– Structurally the same

• Eq? equal? for symbols.

Page 12: More Scheme

Examples

(equal? ‘foo ‘foo) (eq? ‘foo ‘foo) true

(equal? ‘(a b) ‘(a b)) true

(eq? ‘(a b) ‘(a b)) false

(define x ‘(a b c))(define y (cons (car x) (cdr x)))(equal? x y)

true(eq? x y)

false

(define mylist ‘(a b c))(eq? (member ‘b mylist) (cdr mylist))

true

Page 13: More Scheme

More List Functions

• append takes two arguments and returns the concatenation of two lists. be careful here not to confuse append with cons.

(append '(a b c)  '(d e f))  (a b c d e f) (append '(a b (c))  '((d) e f))   (a b (c) (d) e f)

• list returns a list constructed from its arguments.  (list 'a) (a) (list 'a 'b 'c 'd 'e 'f) (a b c d e f) (list '(a b c)) ((a b c)) (list '(a b c) '(d e f) '(g h i)) ((a b c)(d e f)(g h i))

Page 14: More Scheme

More List Functions

• length returns the length of a list. (length '(a b c d ef gh i jk)) 8 (length '(a b c d (ef gh i jk))) 5

• reverse returns the same list, only in reversed order. Note, this is only shallow reverse.

(reverse '(a b d g o)) (o g d b a) (reverse '(a (b d) g o))   (o g (b d) a)

Page 15: More Scheme

Exercise

• Write a helper function to return the first half of a list. Here is the main function:

• Write a helper function to return the second half of a list. Here is the main function:

(define (firsthalf lst) (getfirsthalf lst (quotient (length lst) 2)))

(define (secondhalf lst) (getsecondhalf lst (quotient (length lst) 2)))

Page 16: More Scheme

Merge Method

• Write a method called Merge that merges two sorted lists:(define (Merge x y)

)

Page 17: More Scheme

MergeSort

• Write a MergeSort method that uses your Merge method to sort a list of numbers.

Page 18: More Scheme

let & let*

(let ((x1 E1)

(x2 E2)

(x3 E3)

….

(xn En))

F)

- Expressions E1, E2, … En are evaluated

- Evaluate F with xi’s bound to Ei’s

- Value of let is the value of F

Page 19: More Scheme

Local Variables(let & let*)

• Used to factor out common expressions

• Introduce names in subexpressions

• Order of evaluation of expression is undetermined (let)

• Order of evaluation of expression is sequential (let*)

Page 20: More Scheme

Examples

(let ((x 2)

(y x))

y)

0

(let* ((x 2)

(y x))

y)

2

Page 21: More Scheme

Tail recursion

A recursive function is tail-recursive if

(a) it returns a value without needing recursion OR

(b) simply the result of a recursive activation

i.e. just return the value at the end

• Can be efficiently implemented– Don’t need stacks

• Can convert many functions to be tail recursive

Page 22: More Scheme

Examples

(define factorial

(lambda(n)

(cond ((= n 1) 1)

(else (* n (factorial (- n 1)))))))

Factorial Function

Page 23: More Scheme

ExamplesFactorial Function: Tail Recursive

(define factorial2

(lambda(n m)

(cond ((= n 1) m)

(else (factorial2

(- n 1)

(* m n))))))

(define factorial

(lambda(n)

(factorial2 n 1)))

Page 24: More Scheme

Examples

Getridof Function (get rid of an item from a list)

(define getridof

(lambda(list item)

(cond ((null? list) '())

((equal? item (car list)) (getridof (cdr list) item))

(else (cons (car list) (getridof (cdr list) item))))))

Page 25: More Scheme

Examplesgetridof Function (Tail recursive)

In reverse order! Could you put in original order?

(define gro

(lambda(list item list2)

(cond ((null? list) list2)

((equal? item (car list)) (gro (cdr list) item list2))

(else (gro (cdr list) item (cons (car list) list2))))))

(gro '(x y x z x w) 'x '())

(w z y)

Page 26: More Scheme

Functions

• Functions are first class citizens in Scheme– Variables may be bound to functions

– Can be passed as parameters

– Can be returned as values of functions

Page 27: More Scheme

Functions as First Class Citizens

A function may be bound to a variable, we are already doing this:

(define add1 (lambda(x) (+ 1 x)))

(add1 4)

5

Page 28: More Scheme

Functions as First Class Citizens

Functions can be passed as parameters

(define (foo x y) (x y)

)

(foo (lambda(x) (+ 1 x)) 4)5

Page 29: More Scheme

Functions as First Class Citizens

Functions can be returned as values of functions

(define (foo x) (lambda(y) (+ x y)))

(foo 4)

#<procedure:2:16>

((foo 4) 5)9

Page 30: More Scheme

Examples(map)

(map cdr ‘((1 2) (3 4)) ((2) (4))(map car '((a b) (c d) (e f) (g h)))(a c e g)

• Takes two arguments– Function and a list– Applies function to the list

Page 31: More Scheme

Examples(map)

(let

((proc (lambda(ls) (cons 'a ls))))

(map proc '((b c) (d e) (f g h))))

((a b c) (a d e) (a f g h))

Page 32: More Scheme

Apply

(apply + ‘(4 11))

15

(apply max ‘(3 4 5))

5

(apply <function> <arguments-in-a-list>)

Useful when the arguments are built separately from application

Page 33: More Scheme

Eval

• Eval will evaluate the parameter as a valid Scheme expression

(eval ‘(car ‘(a b c))a

(eval ‘(define (foo a b) (+ a b)))foo

(foo 3 4)7

Allows for interesting opportunities for code to modify itself andexecute self-generated code

Page 34: More Scheme

Binding of Variables

• Global binding– define

• Local binding– lambda, let

• How do we change the value of a variable to which it is bound?– We have been using define multiple times, although in

some implementations of Scheme this is invalid

Page 35: More Scheme

set!

(set! var val)

• Evaluate val and bind it to var• ! Indicates a side effect

• Scheme does not specify what this returns– Implementation dependent

– DrScheme seems to return nothing

Page 36: More Scheme

Examples

(define f (lambda(x) (+ x 10)))

(f 5)

>15

(set! f (lambda(x) (* x 10)))

(f 5)

>50

Page 37: More Scheme

Examples

(let ((f (lambda(x) (+ x 100)))) (display (f 5)) (newline) (set! f (lambda(x) (* x 100))) (f 5))

105 500(f 5) Error, reference to undefined identifier: f

Scheme uses lexical scoping.

Page 38: More Scheme

set-car! & set-cdr!

(define x '(4 5 6))

(set-car! x 7)

x

> (7 5 6)

(set-cdr! x '( 7 8 9))

x

>(7 7 8 9)

Page 39: More Scheme

More Examples

(define (my-reverse ls)

(cond ((null? ls) '())

(else (append (my-reverse (cdr ls))

(list (car ls))))))

(my-reverse '(a b c d))

Page 40: More Scheme

Recursive Functions

(define (atom? x) (not (list? x)))

(define (super-reverse ls)

(cond ((null? ls) '())

((atom? ls) ls)

(else (append (super-reverse (cdr ls))

(list (super-reverse (car ls)))))))

(super-reverse '((a b) ((c d) e) f))

Page 41: More Scheme

Recursive Functions

(define (pairup x y) (cond ((null? x) '()) ((null? y) '()) (else (cons (list (car x) (car y)) (pairup (cdr x) (cdr y))))))

(pairup '(a b c) '(1 2 3))

Page 42: More Scheme

Recursive Functions

(define (listify ls)

(cond ((null? ls) '())

(else (cons (list (car ls))

(listify (cdr ls))))))

(listify '(1 2 2 3))

Page 43: More Scheme

Summary

• Pure functional programming• No assignments (side effects)• Refreshingly simple • Surprisingly powerful

– Recursion- Functions as first class objects

• Implicit storage management– Garbage Collection